In vivo-induced argininosuccinate lyase plays a role in the replication of Brucella abortus

In vivo-induced argininosuccinate lyase plays a role in the replication of Brucella abortus in RAW264.7 cellsXiangan Han,Xiaoqing Sun,Xueqing Shan,Min Zhang,Jun Song, Mingxing Tian,Guobo Fan,Shaohui Wang,Yongliang Tong,Chan Ding and Shengqing YuCorrespondence Shengqing Yuyus@Received1September2013 Accepted8January2014Shanghai Veterinary Research Institute,Chinese Academy of Agricultural Sciences(CAAS),518Ziyue Road,Shanghai200241,PR ChinaBrucellosis caused by Brucella species is a zoonotic disease with a serious impact on public health and the livestock industry.To better understand the pathogenesis of the disease,invivo-induced antigen technology(IVIAT)was used to investigate the in vivo-induced antigens of Brucella abortus in this study.A genomic expression library of B.abortus was constructed and screened using pooled bovine B.abortus-positive sera by IVIAT.In total,33antigens were identified.Five antigens were further expressed and tested for their seroreactivity against33 individual bovine B.abortus-positive sera by Western blot analysis.The results showed a highest positive rate of32/33for argininosuccinate lyase(ASL),indicating that ASL may be used as a candidate marker for serodiagnosis of brucellosis.Furthermore,an asl gene-deleted mutant strain S2308D ASL was constructed,and the intracellular survival and replication of the mutant strain in RAW264.7cells were investigated.Interestingly,the numbers of bacteria recovered from cells infected with mutant strain S2308D ASL were similar at all time points observed from0h to96h post-infection,suggesting the asl gene plays an important role in the bacterial replication in RAW264.7cells.Real-time quantitative PCR(qPCR)analysis showed that the mRNA levels in S2308D ASL were decreased for BvrR,BvrS and virB5when compared with those in S2308 (P,0.05).Our results not only expand the knowledge of Brucella intracellular replication but also expand the list of candidates for serodiagnostic markers of brucellosis.INTRODUCTIONBrucella abortus is a Gram-negative,intracellular bacterium that can survive within a variety of cells and maintain a long-lasting interaction with the host cells(Han et al., 2012).Compared with other bacterial pathogens,B.abortus does not display classical virulence factors such as exo-toxins,capsule,toxic LPS or plasmids.The virulence of B. abortus depends on its properties of survival and replica-tion in host cells(Seleem et al.,2008).It is generally accepted that bacterial genes induced specifically during infection may be essential for bacterial survival and contribute to the pathogenesis of a disease.Therefore,an understanding of the genes involved in B.abortus–host interactions in vivo is crucial for understanding the patho-genesis of brucellosis.Recently,in vivo-induced antigen technology(IVIAT)has been developed to identify immunogenic bacterial genes expressed specifically during infection at the genomic level using sera from infected animals;this technique avoids the limitations of animal models(Gu et al.,2009;Hang et al.,2003;Harris et al., 2006;Kumar et al.,2011;Rollins et al.,2005,2008; Zygmunt et al.,2006).In this study,IVIAT was used to identify B.abortus antigens induced in vivo,in which pooled bovine B.abortus-positive sera were used for the screening.In addition,selected antigens were expressed,and the seroreactivity with33 individual bovine B.abortus-positive sera was tested to evaluate their possibility as candidate markers for serodiag-nosis.Furthermore,an asl gene-deleted mutant strain S2308D ASL was constructed to investigate the role of the gene on the pathogenesis of B.abortus.METHODSBacterial strains,plasmids and growth conditions.B.abortus strain2308(a virulent wild-type strain)was obtained from the Chinese Veterinary Culture Collection Center and cultured in TrypticAbbreviations:ASL,argininosuccinate lyase;IVIAT,in vivo-inducedantigen technology;qPCR,quantitative PCR.Microbiology(2014),160,567–575DOI10.1099/mic.0.072926-0 072926G2014SGM Printed in Great Britain567Soy Agar(TSA;Difco)or Tryptic Soy Broth(TSB;Difco)at37u C with5%CO2.Escherichia coli strains DH5a(Invitrogen)and BL21 (DE3;Stratagene)were cultured at37u C in Luria–Bertani(LB) medium containing50m g kanamycin ml21or100m g ampicillin ml21 as needed.The expression vectors pET-28a,pET-28b,and pET-28c were purchased from Novagen.Restriction enzymes were obtained from MBI Fermentas.Horseradish peroxidase(HRP)-labelled anti-bovine IgG was from KPL.All the chemicals used in this study were of analytical grade and purchased from Sigma.Preparation of the bovine B.abortus-positive sera.To obtain sera that reacted only with the B.abortus antigens expressed during infection,we pooled eight sera from cattle that were naturally infected with B.abortus.The pooled sera were adsorbed extensively with in vitro-grown B.abortus and E.coli BL21(DE3)to remove antibodies that reacted with the bacterial antigens expressed during culture in vitro,as described previously(Harris et al.,2006).Briefly,500m l pooled serum was subjected to four successive direct absorptions with B.abortus cells and cell lysates(prepared by three cycles of freezing and thawing,followed by sonication)and then four cycles with BL21(DE3)cells(containing pET28a/b/c plasmids)and lysates.Each adsorption was performed at4u C on a rocking platform for3h.To evaluate the efficacy of adsorption,a small aliquot(10m l)of pooled sera was saved after each adsorption step,and the OD450values of the sera were detected using an ELISA reaction against either whole bacterial cells of heat-inactivated B.abortus S2308(80u C for30min) and E.coli BL21(DE3),or their cell lysates.The adsorbed sera were used for screening of the expression library.Construction of the expression library and identification of the genes induced during infection by IVIAT.An expression library was constructed with the pET-28abc series of expression vectors.The expression vectors were digested with the restriction enzyme Bam HI (New England Biolabs),gel purified using the GeneJET PCR Purification kit(Fermentas)and treated with shrimp alkaline phosphatase(TaKaRa)for use in the study.Genomic DNA of B. abortus S2308was digested with the restriction enzyme Sau3AI.The DNA fragments ranging from500to3000bp were excised and purified using the GeneJET Gel Extraction kit(Fermentas).The purified genomic DNA was ligated with vector DNA,and subse-quently transformed into competent E.coli BL21(DE3)for construc-tion of the expression library.To expose the induced antigens,the bacterial colonies were lysed by saturating discs with chloroform vapour for15min and air-drying. This was followed by flow-washing and blocking of the nitrocellulose filter membranes,and the membranes were incubated with the adsorbed sera at a dilution of1:1000in PBS–Tween20(PBST)at room temperature for2h.The membrane was washed and incubated with HRP-labelled affinity-purified antibody against bovine IgG (1:3000;KPL)in PBST at room temperature for1h.The membrane was washed and subsequently developed with3,39-diaminobenzidine (DAB;Sigma)until optimum colour development was observed. Plasmid DNA from positive clones was extracted according to standard protocols(Sambrook&Russell,2011)and sequenced on a PRISM DNA Analyser3730(Applied Biosystems)using specific primers T7P:59-TAATACGACTCACTATAGGG-39and T7T:59-GCTAGTTATTGCTCAGCGG-39.The homologies,protein families, and putative functions of the encoded proteins were analysed with BLAST program in NCBI(/pubmed/). The cellular localizations of these proteins were predicted using PSORTb version3.0.2(/psortb/).Expression and seroreactivity of five selected antigens.To investigate whether the gene products identified could be used as diagnostic markers for brucellosis,five genes,including argininosuc-cinate lyase(ASL),short-chain dehydrogenase/reductase(SDR),nitrite reductase family protein(Nitrite),fumarate hydratase class I (Fumhy),and transcriptional regulatory protein(Trp),were selected for expression.The genes were amplified from the B.abortus genome by PCR using the respective primer pairs(Table1)and cloned into pET28a(+)(Novagen).The plasmids were introduced into E.coli BL21(DE3)for IPTG-inducible expression of recombinant proteins and analysed by SDS-PAGE and Coomassie blue staining.The recombinant proteins were purified using HisTrap chelating high-performance columns(Amersham Pharmacia Biotech)and were assayed quantitatively using a bicinchoninic acid(BCA)protein assay kit(Pierce).Western blot was used to determine the seroreactivity of expressed antigens with bovine sera as described(Han et al.,2012).A total of33 bovine B.abortus-positive sera were used,which were determined to be positive for B.abortus infection by a government-approved brucellosis testing method carried out at the Shanghai Center for Animal Disease Control and Prevention in China.A serum sample from a healthy bovine animal was used as the negative control(Han et al.,2012).The expressed recombinant proteins were subjected to SDS-PAGE,transferred onto PVDF membranes(Amersham Pharmacia Biotech)and incubated with33bovine B.abortus-positive sera(1:1000)at room temperature for2h.The membranes were washed,incubated with HRP-labelled affinity-purified antibody against bovine IgG(1:3000;KPL)at room temperature for1h, and subsequently washed and developed with DAB until optimum colour development was observed.Construction of mutant strain S2308D ASL.An asl gene-deleted mutant strain S2308D ASL was constructed by allelic exchange,which involved insertion of a chloramphenicol resistance cassette(1056bp) in a Bam HI site,as described previously(Zhang et al.,2013).Briefly, the upstream and downstream fragments of the asl gene of B.abortus S2308(GenBank accession number NC_007618)were amplified by PCR using the primer pairs ASL-upF/ASL-upR and ASL-downF/ ASL-downR(Table1).The fragments were then used as templates for a second round of PCR using primers ASL-upF and ASL-downR. The resulting fragment was inserted into a pUC19-sacB plasmid at the Xba I site to generate a pUC19-SacB-asl plasmid.A chloram-phenicol resistance cassette,obtained from pR326(TaKaRa)using the primers cat-F/cat-R with Bam HI digestion,was subsequently inserted into the recombinant plasmid Up-pUC19-Down to form pUC19-ASL-cat,which was transformed into B.abortus S2308 competent cells.After the electric pulse,the cells were plated on TSA plates containing5m g chloramphenicol ml21.After72h incubation, the resulting chloramphenicol-resistant colonies were selected for PCR analysis of chromosomal DNA.The mutant with the inactivated asl gene was verified by PCR analysis with primers ASL-inF/ASL-inR and named S2308D ASL.The mutant strain was also identified by Western blot using rabbit antiserum against His-ASL.Growth curves in TSB medium and bacterial persistence in RAW264.7murine macrophage cells(RAW264.7cells). RAW264.7cells were obtained from the ATCC and grown in Dulbecco’s modified eagle medium(DMEM)supplemented with 10%FBS and antibiotics at37u C with5%CO2.The growth curves in TSB medium and bacterial persistence in RAW264.7cells of S2308D ASL and S2308were monitored as described previously (Zhang et al.,2013).The OD600values were monitored at4h intervals for72h to determine the growth rate.The bacterial persistence was determined on RAW264.7cells by infection of S2308D ASL or S2308at an m.o.i.of100.Colony-forming units were counted at12h intervals from0h to96h post-infection.The experiment was performed in triplicate,and the mean c.f.u.at each time point was used to evaluate the persistence of the bacteria in the cells.X.Han and others568Microbiology160Real-time quantitative PCR(qPCR).The transcriptional levels of eight virulence-related genes virB4,per,sodC,virB8,omp25,BvrR, BvrS and virB5in S2308D ASL were determined using qPCR as described previously(Han et al.,2013).Total RNA of the bacteria was extracted using Trizol reagent(Invitrogen)according to the manu-facturer’s protocol.The cDNA was synthesized with total RNA(2m g, A260/A28051.8–2.0),hexa-deoxyribonucleotide mixture random primers(TaKaRa)and M-MLV reverse transcriptase(Promega)as described previously(Han et al.,2013).qPCR was performed using a realplex quantitative PCR system(Eppendorf)in a total volume of 20m l containing10m l of PCR Master Mix(SYBR Green PCR Master Mix;Promega),1m l of cDNA(or water as a negative control),0.5m l of10m M PCR forward primer,0.5m l of10m M PCR reverse primer (Table1),and8m l water.qPCR was performed with an initial denaturation step of10min at95u C,followed by40cycles of15s at 95u C,and1min annealing/extension at60u C.Total cDNA abun-dance in the test samples was normalized using the glyceraldehyde-3-phosphate dehydrogenase(GAPDH)gene as a housekeeping control. The relative change in gene expression was recorded as the ratio of normalized target concentrations.At the end of each cycle,the fluorescence emitted by SYBR Green was measured.Amplification of a single expected product was confirmed by melting curve analysis and electrophoresis in1%agarose gels.Statistical analysis.The statistical significance of the differences between mean values was determined using Student’s t-test(P,0.05). RESULTSAdsorption of pooled bovine B.abortus-positive seraThe sera from eight cattle infected with B.abortus were pooled,and serum adsorption was carried out using inacti-vated whole bacterial cells and cell lysates of in vitro-grown B.abortus S2308and E.coli BL21(DE3),respectively.The efficiency of adsorption was determined by examining the immunoreactivity of the serum aliquots from the pooled bovine sera after each round of adsorption.As shown in Fig.1,the immunoreactivity of the sera with in vitro-grownTable1.Primers used in this studyName Oligonucleotide sequence(5§to3§)Product ID Product size(bp) SDR-F CGC GGATCC ATGACAATTCAGAAAGTCGC a*BAB1_0736705SDR-R CCC AAGCTT TCAGACTGAGCGTGTCAGG bAsly-F CGC GGATCC ATGAGCGAACAAAAATCAAG BAB1_19821401Asly-R CCC AAGCTT TCAGGCCTTGGCAATACGFumhy-F C GAGCTC ATGGCAGAAGCAAGCACC c BAB1_09771620Fumhy-R GC GTCGAC TCAACTCAGGTTGAATTCCTTTrp-F CGC GGATCC ATGAGATTGCCCCTTCCAC BAB2_07631425Trp-R CCC AAGCTT TTATCTCGGCATAAGTCTCGAANitrite-F CG GGATCC ATGTTGAACCAGCAAACCAT BAB1_13041230Nitrite-R CG GAATTC TTATTGATGTCGGGCGTTC dASL-upF ACAT GCATGC GCTCGAAAAAGGTGCATCGGTGA e1180ASL-upR CG GGATCC TTCTGCCGCAATGATGCCCGASL-downF CG GGATCC ATCTGGCAAAGCTCTCGCTTGA1179ASL-downF GG GGTACC TCGGCCTGCAATTCCTTCAC fcat-F CGC GGATCC AATTCGATGGGTTCCGAGG1056cat-R CGC GGATCC ACCGAACTAGAGCTTGATGASL-inF CAAGGAAATAGAAGAGGGCAAAT430DASL-inR TGTGACGGTCGATGGGAAAvirB5-F CTGCGTAGCTATCTGCCACA BAB2_0064227virB5-R GGACGTCATGACCTTGTCCTBvrR-F CAACATCCTGACCTCCGTTT BAB1_2092186BvrR-R ATCGATCTCATCGTCCTTGGBvrS-F TCGCAGAGACCCAGAAAGAT BAB1_2093156BvrS-R TAAAGAATGCCGGAAACCAGsodC-F AGAAAAAGACGGCAAGATCG BAB2_0535187sodC-R CTTGAGATGTGGAGCGACAAper-F GACGATTCCGGCTTCTGTAA BAB1_0541210per-R GGTCTCGGACGTTAAACCAAomp25-F AAGTCAAGCAGGGCTTTGAA BAB1_0722232omp25-R TGTTGCCGTACTGGGTGTAAGAPDH-F CATTCAGGTCGTCGCCATCA BAB1_1741185 GAPDH-R CTTCCTTCCACGGCAGTTCG*The following restriction sites are underlined:a,Bam HI;b,HindIII;c,Sac I;d,Eco RI;e,Sph I;f,Kpn I.D The size of the mutant was zero,while that of the wild strain was430bp,when using the primers ASL-inF/ASL-inR.The roles of Brucella abortus argininosuccinate lyase 569B.abortus and E.coli BL21(DE3)decreased progressively with each round of adsorption;the decrease in immuno-reactivity was particularly noticeable after the first adsorption step.The OD 450value of the serum reacted with B.abortus cells decreased from 2.7to 0.43after four successive adsorption steps.Construction of the expression libraryThe genomic DNA was digested by different concentra-tions of Sau 3AI,of which the most suitable concentration was 0.07U (m g DNA)21(Fig.2a).The DNA fragments of 500–3000bp were ligated into digested pET28-abc,and the optimum ratio of vector to fragments was 2:1.The B.abortus expression library contained approximately 90000colonies.To determine the percentage of transformants containing inserts,the library was plated on LB–kanamycin plates,and 50colonies were picked randomly and analysed by PCR.More than 90%of the transformants contained inserts ranging from 500to 3000bp (Fig.2b).Identification of the gene products induced during infectionThe 90000clones from the inducible B.abortus expression library in E.coli BL21(DE3)were immunoscreened with adsorption-treated pooled bovine sera.In total,33genes were identified;their encoded proteins were classified into different categories and listed in Table 2.The cellular localizations of the proteins were also predicted (/)and are shown in Table 2.Expression and seroreactivity of the selected proteinsRecombinant proteins of ASL,SDR,Nitrite,Fumhy and Trp were expressed successfully and tested for their sero-reactivity with 33individual bovine B.abortus -positive sera by Western blot analysis.The positive rates were 32/33,31/33,24/33,18/33and 13/33,respectively.Construction of the mutant strain S2308D ASL The deletion of the asl gene in mutant strain S2308D ASL was confirmed by PCR amplification.No PCR product was shown in S2308D ASL,but a 430bp product was obtained from the wild-type strain S2308using primers ASL-inF/ASL-inR (Fig.3a,lane 1,2).A 1056bp PCR product was amplified from S2308D ASL using primers cat-F/cat-R (Fig.3a,lane 4),which was consistent with the expected size for the chloramphenicol resistance cassette inserted asl gene.However,no product was obtained in the wild-type strain S2308(Fig.3a,lane 3).The deletion of the asl gene was confirmed further by Western blot using rabbit antiserum against His-ASL.As shown in Fig.3b,the mutant strain S2308D ASL showed no band at 52kDa,representing no binding of ASL (Fig.3b,lane 1),while the wild-type strain S2308showed a band at the size of 52kDa,representing binding of ASL (Fig.3b,lane 2).Bacterial growth in TSB medium and in RAW264.7cellsThe growth rate of S2308D ASL in TSB medium was similar to that of S2308within 72h (Fig.4a).When monitoring the bacterial growth in RAW264.7cells at an infection dose of 100m.o.i.,S2308D ASL remained a similar bacterial c.f.u.of around 1.36104c.f.u.per well in RAW264.7cells before 60h post-infection,and a slight increased c.f.u.was monitored at 72h,84h and 96h (around 5.06104c.f.u.per well).While S2308keeps growing in the cells post-infection,it was determined as 8.56103c.f.u.per well at 0h,2.16107c.f.u.per well at 48h,and 7.06107c.f.u.per well at 96h.The c.f.u.of S2308at 96h in the cell culture was over 1400-fold higher than that of S2308D ASL (Fig.4b).Expression profiles of the virulence-related genes Real-time qPCR results showed that the mRNA levels of BvrR,BvrS and virB5were decreased significantly inAdsorbed with the cell lysates of B. abortusO D 450(a)(b)11.21.4Adsorbed with the cell lysates of E. coli (BL21)E. coli (BL21)0.60.8O D 4500.20.4Adsorbed with the whole cells of Unadsorbed sera Firstadsorption Second adsorption Thirdadsorption Fouth adsorptionUnadsorbedseraFirst adsorptionSecond adsorptionThird adsorptionFouth adsorptionFig. 1.OD 450valus of pooled bovine sera after sequential adsorption.(a)ELISA plates coated with inactivated bacterial whole cells or cell lysates of in vitro -grown B.abortus S2308,respectively.(b)ELISA plates coated with whole bacterial cells or cell lysates of in vitro -grown E.coli BL21(DE3),respectively.Results are expressed as means of absorbance values for three individual experiments,and error bars represent SEM .X.Han and others570Microbiology 160S2308D ASL compared with S2308(P ,0.05).The mRNA levels of per ,sodC and omp25in S2308D ASL showed no significant differences from those in S2308(Table 3).DISCUSSIONBrucellosis is a zoonosis that is distributed worldwide,affects a broad range of mammals,ranging from domestic animals to humans,and causes important economic losses (Olsen &Tatum,2010).An understanding of the patho-genicity of the disease,especially the mechanism of infec-tion in vivo ,is very important for control of the infection.The development of IVIAT has provided a valuable tool for the identification in vivo of genes induced during infection (Rollins et al.,2008).In this study,33gene products were identified using IVIAT with bovine sera.In comparison with a previous study that used elk sera (Lowry et al.,2010),32novel gene products were discovered,with the exception of malate dehydrogenase (MDH).This may be explained as follows:(1)B.abortus modulates its gene expression to adapt to different hosts,resulting in the expression of different gene products during infection.(2)The bovine sera we used were from different areas and processes of infection,which allowed us to identify thewidest possible array of antigens produced during different stages of infection.The analysis of diverse functional classes showed that 33antigens were involved in cell envelope structure,substance and energy metabolism,regulation,molecule synthesis,transport and translation.Previous studies showed that the genes related to metabolism may have functions essential for Brucella growth in vivo ,a feature that is essential for bacterial pathogenicity (Barbier et al.,2011).However,studies have shown that these genes involved in synthesis and metabolism are not necessary when B.abortus is cultured in vitro (Al Dahouk et al.,2010).ASL is an enzyme that catalyses argininosuccinate (ASA)producing arginine and fumarate,and fumarate can enter the citric acid cycle,which is a key component of the metabolic pathway by which all aerobic organisms generate energy.ASL is a key enzyme in the conversion of ammonia to urea through the urea cycle.Ammonia builds to toxic levels,resulting in hyperammonemia (Peuscher et al.,2012).Recent studies suggested that ASL from eukaryotic cells modulates autophagy and cell death in glioblastoma (Syed et al.,2013).To investigate the role of B.abortus asl gene on the pathogenesis,an asl gene-deleted mutant strain S2308D ASL was constructed and the growth in RAW267.4cells was(a)12345MM 1234567891011121314151617181920212223245000 bp 3000 bp 2000 bp 1500 bp 1000 bp 750 bp 500 bp 250 bp 100 bp5000 bp 3000 bp 2000 bp 1500 bp 250 bp 100 bp750 bp 500 bp (b)Fig.2.Construction of the genomic expression library.(a)Optimization of the concentration of Sau3AI for digestion.Genomic DNA (1m g)was digested in 45m l reaction volumes with different concentrations of Sau nes 1to 5contained 0.10U,0.09U,0.08U,0.07U and 0.06U of Sau 3AI,respectively.The products were detected by electrophoresis in 0.8%agarose gels,and 0.07U of Sau 3AI was determined to be a suitable concentration for the digestion of genomic ne M:DL 5000DNA marker (3428A;TaKaRa).(b)Percentage of transformants containing inserts.Fifty colonies were chosen randomly and analysed by qPCR with primers T7P and T7T,which were specific to pET28-abc;47colonies contained genomic DNA inserts of 500–3000bp (94%).The results for 24colonies are presented in the ne M:DL 5000DNA marker (3428A;TaKaRa).The roles of Brucella abortus argininosuccinate lyase571measured.The results showed a significant decrease in replication in RAW264.7compared with the wild-type S2308strain.The c.f.u.of S2308inside the RAW264.7in each well increased5400times from0h to96h,while the c.f.u.of S2308D ASL increased only fourfold.This indicates that S2308D ASL could not replicate in the cells as successfully as S2308.This may be explained by the following reasons:(1)ASL deficiency results in defective cleavage of ASA(Haines et al.,2011).This leads to an accumulation of ASA in cells,which causes hyperammonemia in affected individuals,and may result in significant decrease in replication in RAW264.7.(2)ASL can catalyse ASA to produce fumarate,which is a key component of citric acid cycle by which all aerobic organisms generate energy.ASL deficiency would result in decrease the level of energy generation due to entering fumarate insufficient the citricTable2.In vivo-induced antigens identified by IVIATCategory Product ID Gene product description*Predicted cellsublocalization DMolecular mass(kDa)Cell envelope structure BAB1_2176Lipoprotein YaeC family Cytoplasmic membrane27.9 BAB2_1102Chemotaxis motc protein precursor Unknown40.1 Substance and energymetabolismBAB1_1982Argininosuccinate lyase Cytoplasm51.3BAB1_0781Enoyl-CoA hydratase/isomerase Cytoplasm37.8BAB1_0925Biotin/lipoyl attachment:Acetyl-CoA biotin carboxylcarrier:Biotin-requiring enzyme,attachment siteCytoplasm49.2BAB1_0736Short-chain dehydrogenase/reductase SDR:Glucose/ribitolde hydrogenaseCytoplasm24.9 BAB1_0977Fumarate lyase:Fe–S type hydro-lyasestartrate/fumarate alpha region:Fe–S type hydro-lyasesCytoplasm59.0BAB1_1304Nitrite/sulfite reductase ferredoxin-like half domain Unknown46.3BAB2_03463-isopropylmalate dehydrogenase Cytoplasm39.7BAB1_1927Malate Dehydrogenase Cytoplasm35.0 Regulation BAB2_0763Response regulator receiver:Transcriptionalregulatory protein,C-terminalCytoplasmic membrane52.5 BAB1_0354Bacterial regulatory protein LysR,HTH motif:LysRsubstrate binding domainCytoplasmic membrane32.3BAB1_1803Bacterial regulatory protein,GntR family Cytoplasm25.3BAB1_1059PAS domain:ATP-binding region,ATPase-like:Histidine kinase ACytoplasmicMembrane98.0BAB2_0220Bacterial regulatory protein,MerR family Cytoplasm17.3BAB1_0354Bacterial regulatory protein LysR Cytoplasmic membrane32.3 Molecule synthesis BAB1_1145Glucose/ribitol dehydrogenase Cytoplasm36.0 BAB1_0711DNA polymerase III chi subunit,HolC Unknown17.2BAB1_0873Beta-ketoacyl synthase Cytoplasmic membrane40.9BAB1_1925CoA binding domain Cytoplasm31.3BAB1_0896Arginyl-tRNA synthetase Cytoplasm65.2BAB1_1573Chaperonin clpA/B Cytoplasm36.6BAB1_2127ATP/GTP-binding site motif A Cytoplasm85.8BAB2_0709ATP/GTP-binding site motif A Cytoplasmic membrane88.8 Transport BAB2_1133Bacterial inner-membrane translocator Cytoplasmic membrane59.6 BAB1_0583Mg2+transporter protein Cytoplasmic membrane36.5BAB2_1142Binding-protein-dependent transport systems,innermembrane componentCytoplasmic membrane34.2 BAB2_0292Sugar transporter Cytoplasm35.3BAB2_0168Autotransporter beta-domain Multiple18.7 Translation BAB1_1257Elongation factor,GTP-binding Cytoplasm42.6 BAB1_1256Ribosomal protein S10Cytoplasm11.6BAB1_1263RNA polymerase,alpha subunit Cytoplasm155.6 Unknown BAB1_1848Protein of unknown function DUF163Cytoplasm18.2*Putative functions of hypothetical proteins were determined from NCBI(/pubmed/).The NCBI database or program was used to determine the gene product description.D The cellular localizations of these proteins were predicted using PSORTb version3.0.2(/psortb/).X.Han and others572Microbiology160acid cycle (Gest,1987).(3)Inactivation of the asl gene may result in downregulation of the transcriptional levels of some genes that mediate replication in RAW264.7,whichwould ultimately decrease S2308D ASL replication in RAW264.7.In this study,Brucella virB5was significantly decreased at the gene transcription level in S2308D ASL,compared with S2308(P ,0.05).The B.abortus virB locus contains 12ORFs,termed virB1through virB12,which encode a type IV secretion system (T4SS)that is required for persistent infection of mice,for survival in tissue culture cells and for inhibition of fusion of Brucella -containing vacuoles with lysosomes (Boschiroli et al.,2002;de Jong et al.,2013;Roux et al.,2007).Furthermore,in the present study,the transcriptional levels of the two-component regulatory system BvrR/BvrS were significantly decreased in S2308D ASL com-pared with S2308(P ,0.05).The two-component regulatory system BvrR/BvrS is essential for B.abortus virulence,and B.abortus BvrR/BvrS controls carbon and nitrogen metabolism,adjusts Brucella physiology during the transit from the extracellular to the intracellular niche (Manterola et al.,2007;Martinez-Nun˜ez et al.,2010;Viadas et al.,2010).Hence,we presumed that the decrease of bacterial replication in RAW264.7caused by deletion of asl gene may be due to downregulation of the transcriptional levels of T4SS virB locus and BvrR/BvrS to a certain degree.Further studies will be designed to elucidate the possible mechanism.In addition to potentially detecting novel in vivo -induced virulence genes,IVIAT has provided the means to identify B.abortus antigenic gene products as markers of infection (Lowry et al.,2010).Classical serological techniques for the(a)2000 bp 95 kDa 72 kDa 55 kDa 43 kDa 34 kDa 25 kDaASL1000 bp 750 bp 500 bp 250 bp 100 bp(b)12345M 123M Fig.3.Construction and identification of asl gene-deleted mutant strain S2308D ASL.(a)PCR analysis of the mutant strain S2308D ne M:DL 2000DNA marker (3427A;TaKaRa).Lane 1:the mutant strain S2308D ASL showed no PCR product using primers ASL-inF/ne 2:the wild-type strain S2308showed a 430bp PCR product using primers ASL-inF /ne 3:the wild-type strain S2308showed no PCR product using primers cat-F/ne 4:the mutant strain S2308D ASL showed a 1056bp PCR product (chloramphenicol resistant cassette)using primers cat-F/ne 5:negative control.(b)Western blot analysis of the mutant strain S2308D ne M:pre-stained protein marker (SM0671,Fermentas).Lane 1:the mutant strain S2308D ASL showed no band for binding to ne 2:the wild strain S2308showed a band representing binding of ASL.(a)(b)O D 600S2308S2308ΔASLl o g (c .f .u . w e l l –1)Fig.4.Bacterial growth curves and persistence in RAW264.7cells.(a)Bacterial growth curves.Growth curves of S2308D ASL and wild-type strain S2308in TSB.OD 600values were monitored at 4h intervals for 72h to determine the growth rate.Data are presented as mean ±SD from three independent experiments.(b)Persistence of S2308D ASL and S2308in RAW264.7cells.RAW264.7cells cultured in 24-well plates were infected with mutant strain S2308D ASL or wild-type strain S2308at an m.o.i.of 100.Bacterial c.f.u.values were determined at 12h intervals for 96h.Data were transformed exponentially and the error bars represent SEM .The roles of Brucella abortus argininosuccinate lyase573。